Physical Solutions of The Symmetric Teleparallel Gravity in Two-Dimensions
Muzaffer Adak
Department of Physics, Faculty of Arts and Sciences, Pamukkale University, 20100 Denizli, Turkey
madak@pau.edu.tr
Tekin Dereli
Department of Physics, Ko¸c University 34450 Sarıyer-Istanbul, Turkey
tdereli@ku.edu.tr
Abstract
A 2D symmetric teleparallel gravity model is given by a generic 4- parameter action that is quadratic in the non-metricity tensor. Vari- ational field equations are derived. For a particular choice of the cou- pling parameters and in a natural gauge, we give a Schwarzschild type solution and an inflationary cosmological solution.
1 Introduction
In order to investigate the physical implications of General Relativity (GR) some simple limiting cases are often sought. The low energy, static limit is attained by assuming the existence of a time-like Killing vector. However, such configurations have no dynamics. Symmetric limits at arbitrary energy scales on the other hand are obtained by assuming one or more space-like Killing vectors. For example, the assumption of spherical symmetry reduces the gravitational action upon integrating over the angular variables to an effective 2D gravity model in (t, r) coordinates. Such 2D gravity models have recently been used much to discuss black hole dynamics, quantized gravity or numerical relativity [1]. It is well known that Einsteinean gravity in 2D has no dynamical degrees of freedom. One way of making this theory dynamical is to couple a dilaton scalar. It is also possible to introduce further dynamical degrees of freedom by going over to a non-Riemannian space-time by the inclusion of torsion and non-metricity. Generalized 2D gravity models with curvature and torsion and teleparallel theories with only torsion have been studied a lot [2, 3]. On the other hand 2D models including non-metricity received less attention [4, 5]. The symmetric teleparallel gravity (STPG) with zero-curvature and zero-torsion has been introduced relatively recently [6, 7, 8]. Here we consider 2D symmetric teleparallel gravity models with the most general 4-parameter action that is quadratic in the non-metricity tensor [9]. For a particular choice of parameters we have shown that the corresponding set of field equations admit both the Schwarzschild solution and an inflationary cosmological solution. It is remarkable that the STPG equivalent of Einstein gravity in 2D does not admit these solutions.
2 Mathematical Preliminaries
The triple {M, g,∇} denotes the space-time where M is a 2-dimensional differentiable manifold, g is a non-degenerate Lorentzian metric and∇is a linear connection. g can be written in terms of the co-frame 1-forms
g=gαβdxα⊗dxβ =ηabea⊗eb (1) where {ea} is an orthonormal co-frame and {dxα} are the co-ordinate co- frame 1−forms. We take ηab = Diag(−,+) and gαβ are the co-ordinate components of the metric. The dual orthonormal frame is determined from the relations eb(Xa) = ıaeb = δba. Similarly, dxβ(∂α) = ıαdxβ = δαβ. Here ı denotes the interior product operator that maps anyp-form to a (p−1)- form. We set space-time orientation by the choice²01= +1 or ∗1 =e0∧e1
where ∗ is the Hodge star. Finally, the connection is specified by a set of connection 1-forms{Λab}. The non-metricity 1-forms, torsion 2-forms and curvature 2-forms are defined through the Cartan structure equations:
Qab = −1
2Dηab= 1
2(Λab+ Λba), (2) Ta = Dea=dea+ Λab∧eb , (3) Rab = DΛab =dΛab+ Λac∧Λcb (4) Bianchi identities follow as their integrability conditions:
DQab = 1
2(Rab+Rba), (5)
DTa = Rab∧eb, (6)
DRab = 0. (7)
If the non-metricity is non-vanishing, special attention is due in lowering and raising indices under covariant exterior derivative. We make use of the identities
D∗ea = −Q∧ ∗ea+²abTb (8)
D∗eab = −²abQ (9)
whereQ= Λaa=Qaais called the Weyl 1-form. The full connection 1-forms can be decomposed uniquely as follows [4, 5]:
Λab =ωab+Kab+qab+Qab (10) whereωab are the Levi-Civita connection 1-forms satisfying
ωab∧eb =−dea, (11) Kab are the contortion 1-forms satisfying
Kab∧eb =Ta, (12)
and qab are the anti-symmetric tensor 1-forms defined by
qab=−(ıaQbc)∧ec+ (ıbQac)∧ec. (13) In this decomposition the symmetric part
Λ(ab) =Qab (14)
while the anti-symmetric part
Λ[ab]=ωab+Kab+qab. (15) In writing down gravity models it easily becomes complicated to keep all the components of Qab. Therefore, people sometimes deal only with certain irreducible parts of that. To obtain the irreducible decomposition of non-metricity tensor under the 2D Lorentz group, firstly we write
Qab= Qab
|{z}
trace−f ree part
+ 1 2ηabQ
| {z }
trace part
. (16)
Then
Qab=(1)Qab+(2)Qab+(3)Qab (17) in terms of
(2)Qab = 1
2[(ıaΛ)eb+ (ıbΛ)ea−ηabΛ], (18)
(3)Qab = 1
2ηabQ , (19)
(1)Qab = Qab−(2)Qab−(3)Qab (20)
where Λ := (ıbQba)ea.These irreducible components satisfy
ηab(1)Qab =ηab(2)Qab= 0, ıa(1)Qab= 0, ea∧(1)Qab = 0. (21) Thus they are orthogonal in the following sense:
(i)Qab∧ ∗(j)Qab =δijNij (no summation overij) (22) whereδij is the Kronecker symbol andNij are certain 2-forms. We calculate
(1)Qab∧ ∗(1)Qab = Qab∧ ∗Qab−(2)Qab∧ ∗(2)Qab−(3)Qab∧ ∗(3)Qab ,
(2)Qab∧ ∗(2)Qab = (ıaQac)(ıbQbc)∗1 +1
4Q∧ ∗Q−(ıaQ)(ıbQab)∗1,
(3)Qab∧ ∗(3)Qab = 1
2Q∧ ∗Q . (23)
3 Symmetric Teleparallel Gravity
We formulate STPG by a variational principle from an action I =
Z
M
³
L+Rabρab+Taλa
´
(24) whereLis a Lagrangian density 2-form quadratic in the non-metricity ten- sor,ρabandλaare the Lagrange multiplier 0-forms imposing the constraints
Rab = 0 , Ta= 0. (25)
The gravitational field equations are derived from (24) by independent vari- ations with respect to the connection {Λab} and the orthonormal co-frame {ea} (and the Lagrange multipliers):
λa∧eb+Dρab =−Σab , (26)
Dλa=−τa (27)
where Σab = ∂Λ∂La
b andτa= ∂e∂La. In principle (26) is solved for the Lagrange multipliersλa andρab and substituted in (27) that governs the dynamics of the gravitational fields. It is important to notice thatDλa rather than the Lagrange multipliers themselves appear in (27). Thus we must be calculating Dλa directly and that we can do by taking the covariant exterior derivative of (26):
Dλa∧eb=−DΣab. (28) We used above the constraints
Ta= 0 , D2ρab=Rbc∧ρac−Rca∧ρcb = 0 (29) where the covariant exterior derivative of a (1,1)-type tensor is given by
Dρab=dρab+ Λbc∧ρac−Λca∧ρcb . (30) Finally we arrive at the gravitational field equations
DΣab−τa∧eb = 0. (31) We now write down the most general Lagrangian density 2-form which is quadratic in the non-metricity tensor [9]:
L= 1 2κ
" 3 X
I=1
kI(I)Qab∧ ∗(I)Qab+k4
³(2)Qab∧eb
´
∧ ∗
³(3)Qac∧ec
´# . (32)
where k1, k2, k3, k4 are dimensionless coupling constants and we also intro- duced κ = 8πGc3 with G being Newton’s gravitational constant. Inserting (23) into (32) we find
L= 1 2κ
h
c1Qab∧ ∗Qab+c2(ıaQac)(ıbQbc)∗1 +c3Q∧ ∗Q+c4(ıaQ)(ıbQab)∗1 i
(33) where we defined new coupling constants:
c1 = k1,
c2 = −k1+k2 , c3 = −3
4k1+1 4k2+1
2k3+1 4k4 , c4 = k1−k2−1
2k4. (34)
We remark at this point that STP equivalent of the Einstein-Hilbert Lagrangian density will be obtained for the choicec1 =−1, c2 = 0, c3 =
−1, c4= 2. We wish to show this briefly. Firstly we use the decomposition of the full connection (10), withKab = 0,
Λab=ωab+ Ωab where Ωab =Qab+qab . (35) By substituting that intoRab(Λ) we decompose the non-Riemannian curva- ture as the follows:
Rab(Λ) = dΛab+ Λac∧Λcb
= Rab(ω) +D(ω)Ωab+ Ωac∧Ωcb. (36) Here Rab(ω) is the Riemannian curvature 2-form and D(ω) is the covari- ant exterior derivative with respect to the Levi-Civita connection. To set Rab(Λ) = 0 for STP space-time yields the Einstein-Hilbert Lagrangian 2- form
LEH = Rab(ω)∧ ∗eab
= −[D(ω)Ωab]∧ ∗eab−Ωac∧Ωcb∧ ∗eab (37) Here after using the equality
d(Ωab∧ ∗eab) = [D(ω)Ωab]∧ ∗eab−Ωab∧[D(ω)∗eab] (38)
we discard the exact form and noticeD(ω)∗eab = 0 because Ta(ω) = 0 and Qab(ω) = 0 (see eq.(9)). Thus, up to a closed form
LEH = 1
2κΩac∧Ωcb∧ ∗eab
= 1
2κ(Qac+qac)∧(Qcb+qcb)∧ ∗eab
= 1
2κ(Qac∧Qcb+qac∧qcb)∧ ∗eab
= 1
2κ[−Qab∧ ∗Qab−Q∧ ∗Q+ 2(ıbQ)(ıaQab)∗1] (39) whereκ is gravitational coupling constant.
4 Solutions
We obtain the following contributions to the variational field equations (31) coming from (33):
Σab = X4
i=0
ciiΣab , τa= X4
i=0
ci iτa (40) where
1Σab = 2∗Qab , (41)
2Σab = ıcQac∗eb+ıcQbc∗ea , (42)
3Σab = 2ηab∗Q , (43)
4Σab = 1
2(ıaQ)∗eb+1
2(ıbQ)∗ea+ηab(ıcQcd)∗ed, (44)
1τa = −(ıaQbc)∧ ∗Qbc−Qbc∧(ıa∗Qbc), (45)
2τa = (ıbQbc)(ıdQcd)∗ea−2(ıaQbd)(ıcQcd)∗eb , (46)
3τa = −(ıaQ)∧ ∗Q−Q∧(ıa∗Q), (47)
4τa = (ıbQ)(ıcQbc)∗ea−(ıaQ)(ıcQbc)∗eb−(ıbQ)(ıaQbc)∗ec.(48) One can consult Ref.[8] for the details of variations.
A generic class of solutions will be obtained in the coordinate frame eα = dxα in which Λαβ = 0. We call this the natural or inertial gauge choice:
Rαβ =dΛαβ+ Λαγ∧Λγβ = 0, (49) Tα =d(dxα) + Λαβ∧dxβ = 0, (50) Qαβ =−1
2Dgαβ =−1
2dgαβ 6= 0. (51)
After a frame transformation via the zweibein ea = haαdxα and setting Λab=haαΛαβhβb+haαdhαb we obtain the orthonormal components
Rab =haαRαβhβb = 0, Ta=haαTα= 0, Qab=Qαβhαahβb 6= 0. (52) This shows that in the natural gauge, the field equations may be solved just by a metric ansatz.
4.1 Static solutions First we consider
g=−f2(r)dt2+g2(r)dr2. (53) The co-ordinate components of the metric and the zweibein reads, respec- tively, as follows
gαβ =
µ −f2 0 0 g2
¶
, haα=
µ f 0 0 g
¶
. (54)
The coordinate components of non-metricity is found to be Qαβ =−1
2dgαβ = Ã f f0
g e1 0 0 −g0e1
!
(55) where prime denotes the derivative with respect to r. The orthonormal components are obtained through a frame transformation
Qab =Qαβhαahβb= Ã f0
f ge1 0 0 −gg20e1
!
. (56)
In the above configuration the only non-trivial field equation comes from the trace of (31):
dΣaa= 0 (57)
that reads explicitly α
"µ f0
f
¶0 +
µf0 f
¶2# +β
"µ g0
g
¶0
− µg0
g
¶2#
+ (β−α)f0 f
g0
g = 0 (58) whereα= 2c1+ 4c3+c4, β = 2c1+ 2c2+ 4c3+ 3c4 . Mathematically, this equation has infinitely many solutions because there are two functions and
only one equation. That is, giveng(r) we determine the correspondingf(r).
Physically, however, due to the observational success of the Schwarzschild solution of GR, we seek solutions of the form
f(r) = µ
a+b r
¶p
, g(r) = µ
a+ b r
¶q
. (59)
Then (58) becomes
(p−q+ 1 + 2ar
b )(pα+qβ) = 0 (60) and therefore we let β = −pqα. We note here that STP Einstein-Hilbert action corresponds to the choice p = 0 with arbitrary q. Finally with p = 1/2, β = α and a suitable choice of a and b, the Schwarzschild metric is obtained:
g=−(1−2m
r )dt2+ dr2
(1− 2mr ). (61)
4.2 Cosmological solutions
For cosmological solutions we start with the metric
g=−dt2+R2(t)dr2 (62)
whereR(t) is the expansion function. In the natural gauge, the non-metricity has just one non-zero component
Q11=−R˙
Re0 , others = 0 (63)
where dot denotes the derivative with respect to t. For this configuration, the only non-trivial part of (31) is, again, its trace;dΣaa= 0 which reads
α
ÃR˙
R
!. +
ÃR˙ R
!2
= 0. (64)
This equation accepts two classes of solution, as well.
1. Forα6= 0, we obtain
R(t) =a+bt (65)
whereaandbintegration constants. The case of STP Einstein-Hilbert action belongs here.
2. Forα= 2c1+ 4c3+c4 = 2k3+ 12k4 = 0, we set ÃR˙
R
!. +
ÃR˙ R
!2
=S(t) (66)
where S(t) is any t dependent function. This result contains many familiar solutions. For example;
• ForS =H2: Hubble constant, we obtain the Robertson-Walker metric satisfying the perfect cosmological principle. A co-ordinate transformation tan√2kr= √2kρ, takes the metric to
ds2=−dt2+e2Ht dρ2
(1 + 14kρ2)2 . (67)
5 Conclusion
In this paper we investigated the symmetric teleparallel gravity in two- dimensions. After giving the orthogonal, irreducible decomposition of the non-metricity tensor under the 2D Lorentz group, we wrote down the most general four-parameter Lagrangian density 2-form quadratic in the non- metricity tensor. We obtained the variational field equations and for any particular choice of coupling constants, we have shown that the correspond- ing field equations in a natural gauge admit both the static Schwarzschild solution and an inflationary cosmological solution.
References
[1] D Grumiller and R Meyer, Ramifications of lineland (arXiv:hep- th/0604049) to appear in The Proceedings of 5th Workshop on Quanti- zation, Dualities and Integrable Systems, 23-28 January 2006, Denizli, Turkey and the references therein.
[2] Y N Obukhov,Phys. Rev.D69 (2004) 064009 [3] M Adak,Gen. Rel. Grav.(arXiv:gr-qc/0509010)
[4] T Dereli and R W Tucker,Class. Quant. Grav. 11(1994) 2575
[5] F W Hehl, J D McCrea, E W Mielke and Y Ne’eman,Phys. Rep.258 (1995) 1
[6] J M Nester and H J Yo, Chinese J. Phys. 37 (1999) 113 (arXiv:gr- qc/9809049)
[7] M Adak and ¨O Sert,Turk. J. Phys. 29(2005) 1 (arXiv:gr-qc/0412007) [8] M Adak, M Kalay and ¨O Sert, Int. J. Mod. Phys. D15 (2006)...
(arXiv:gr-qc/0505025)
[9] Y N Obukhov, E J Vlachynsky, W Esser and F W Hehl, Phys. Rev.
D56 (1997) 7769